On the design of LDPC code ensembles for BIAWGN channels

Existing design methods for irregular Low-Density Parity-Check (LDPC) codes over the additive white Gaussian noise channel are based on using asymptotic analysis tools such as density evolution in an optimization process. Such a process is computationally expensive particularly when a large number o...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on communications 2010-05, Vol.58 (5), p.1376-1382
Main Authors: Saeedi, Hamid, Banihashemi, Amir H
Format: Article
Language:English
Subjects:
Additive white noise
Algorithms
AWGN channel
AWGN channels
code design
Codes
Design
Design methodology
Design optimization
irregular LDPC codes
Iterative algorithms
Iterative decoding
LDPC codes
Low-density parity-check codes
Optimization methods
Parity check codes
Stability
stability condition
Studies
systematic design
Upper bound
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Existing design methods for irregular Low-Density Parity-Check (LDPC) codes over the additive white Gaussian noise channel are based on using asymptotic analysis tools such as density evolution in an optimization process. Such a process is computationally expensive particularly when a large number of constituent variable node degrees are involved in the design. In this paper, we propose a systematic approach for the design of irregular LDPC codes. The proposed method, which is based on a pre-computed upper bound on the fraction of edges connected to variable nodes of degree 3, is considerably less complex than the conventional optimization approach. Through a number of examples, we demonstrate that using our method, ensembles with performance very close to those devised based on optimization, can be designed. In addition to having very good performance, the number of constituent variable node degrees in the designed ensembles is only three or four. This, in some cases, is much smaller than the corresponding number for optimization-based designs with similar performance.
ISSN:0090-6778
1558-0857
DOI:10.1109/TCOMM.2010.05.080299